MALES

In satellite to ground optical communications, turbulence disturbs the optical beam. The vertical profile of the
turbulence strength is needed to understand the relation to the communications performance. The goal of this project
is to develop a machine learning tool to measure the turbulence profile, using measurements of a communications
beam. Conventional methods resort on light from binary stars, which statistics over time are used in the weak
turbulence regime. What we developed instead, is a machine learning approach, that can work with a single satellite
source, deal with more versatile turbulence conditions, at a high temporal resolution, by utilizing the spatial statistics.
Using this method, we estimate the coherence length of the turbulence with a 10% accuracy. Currently we are in the
process of publishing the code in the Journal of Open Source Software and preparing a journal article for the Journal of
Astronomical Telescopes, Instruments and Systems. In discussions with research institutes and agencies, we know this
work is highly interesting for them. With this profiling tool, we plan to do characterization over an optical
communications link from a satellite to a ground station, and hope this will become a standard tool in an optical
ground station.